Hacker news

  • Top
  • New
  • Past
  • Ask
  • Show
  • Jobs

Tencent Hy4 Flash (https://hy.tencent.ai)

174 points by lyra1337 3 days ago | 27 comments | View on ycombinator

chriddyp 2 days ago |

This one scored very well on our data analytics benchmarks [1].

Beat out GPT-6 Luna on the low end of cheap + accurate. Only cost $0.13 to answer 43 data analytics questions (compared to ~$20 on the high end of expensive frontier models).

Once the open the weights, Hy4 Flash and regular Hy4 will be the only two open models on the frontier in the exam I've been running.

[1] https://plotly.com/blog/hy4-flash-plotly-data-analytics-benc...

jauntywundrkind 2 days ago |

Benchmarks are competitive with glm-5.3 flash, and man! That price! Ridiculously cheap. Engrams really rocking it, eh?

224b params. 320b for glm-5.3-flash. 180b for qwen3.8 flash next. 552b for DeepSeek flash v4.1 but that has way stronger performance, for 4x+ the price.

Also, they do mention at the end that yes they are going to open the weights here.

ddxv 2 days ago |

Cheaper pricing than Deepseek 1017 and Luna and supposedly 'faster' (though I guess in practice a lot of that comes down to where it's running.

Lately I let myself drift into using Terra and some other mildly 'expensive' models, but really I think the sweet spot for me is cheap and fast.

thesurlydev 2 days ago |

I've been using this model via openrouter for the last couple of days in place of deepseek flash with good results.

DeathArrow 2 days ago |

There is also this cheap one from Baidu worth checking out: https://ernie.baidu.com/release/ernie-5-5-flash-preview/

tkslucas 2 days ago |

Lovely, it's always great to see new model drops

syntaxing 2 days ago |

Wow 224B with 64B anagrams. It might be possible to run this well on two 128GB machine. I’ve been exclusively using deepseek flash 1017 as my “smart” model through open router. Really curious how this compares

wsxiaoys 2 days ago |

it's doing ok on my multimodal/3D task: https://sketchagent.dev/s/Ss6HgYEq reasoning looks solid and properly accept visual feedback.

More results: https://sketchagent.dev/#/a/hy4-flash

DeathArrow 2 days ago |

I don't think it beats Deepseek v4.1 Flash 1017 at cost per task. Deepseek is really great and there are some really cheap providers if you look for them.

Even a $10 Open Code plan will get you far because it will hit cache a lot if you use a matching harness.

Rodmine 2 days ago |

The smaller size seems like a win. While nothing beats Deepseek 4.1 flash on price/performance, I hope the smaller size means that 3rd party providers can provide much faster speeds (DS 4.1 flash is often available around 200-400 tps in Openrouter), and hence there will be actual competition.

Kuyawa 2 days ago |

I love the smell of fresh competition, it brings prices down and features up, that's the essence of free markets, the real reason behind the "War on AI"

throwa356262 1 day ago |

You would assume the company behind tiktok has enough compute to serve this but their inference speed and reliability has not been the best so far:

Eden reports 40tps for hy4: https://www.edenai.co/models/hy4

And error rate is not insignificant according to Openrouter: https://openrouter.ai/tencent/hy4

ketzu 2 days ago |

Curious to test it, but Hy4 via openrouter was such a nightmare in reliability for me when testing it for basic error correction. A niche usecase, but unlike most models I had so many broken responses there.

Maybe Hy4 flash will at least not also be expensive while breaking on me.

edtechdev 2 days ago |

I thought I could use it for some translation work, but it made many more errors than DeepSeek 4.1 Flash 1017

trilogic 2 days ago |

Interesting claims, very low cost according to the tables. On my way to test it.

vcryan 2 days ago |

I love it. It's a bit slow. Hopefully other providers will pick it up.

undefined 2 days ago |

undefined

undefined 3 days ago |

undefined

paramseast 2 days ago |

[dead]

TheJCDenton 2 days ago |

[dead]

scottchiefbaker 2 days ago |

[dead]

italiansolider 2 days ago |

[dead]